¹ Let us say that for any being B, B is wholly good just in case B is a conscious agent who never does evil. As will soon become evident, this paper's main line of argument can accommodate any other plausible characterization of a wholly good agent, as long as this characterization is uniformly applied to those agents involved in the comparison.Google Scholar

² See, for example, Gulesarian (1983) and Pike (1969).Google Scholar

³ This paper's central argument does not depend upon the candidates for God, namely G and G*, having a specific material status as long as both G and G* are posited to have the same material status in any given comparison (since, clearly, there are action-types that a material being could perform, but that a non-material being could not). I shall, for the sake of convenience, and in order to be consistent with the orthodox Christian conception of God, stipulate that G and G* are immaterial beings.Google Scholar

⁴ Because of set-theoretic considerations first pointed out by Bertrand Russell, S is not to be identified with ‘the set of all propositions’.Google Scholar

⁵ This may not be perspicuous to one who individuates propositions differently for God than for us (for example, in accordance with the medieval doctrine of divine simplicity).Google Scholar

⁶ To keep things simple, let us say that g knows the truth-values of neither (7) nor (7′), but g does know the truth-value of the conjunctive proposition: ‘Poe wrote “Berenice as conjuncts.Google Scholar

⁷ x201D; and Griswold loathed Poe,’ and any other conjunctive proposition including (7) or (7′)The same simplifying assumptions made in note 6 are applicable here.Google Scholar

⁸ An Augustinian platonist might claim that only God exists in W, while theistic platonists of other stripes would insist that God, along with all other necessary objects, exists in W.Google Scholar

⁹ Note that St Thomas would deny that there is any possible world in which God is ‘inactive’ (in virtue of the eternal active love which necessarily obtains between the three persons of the Trinity). Aquinas' point is well appreciated. But, I am not here implying that God does not perform any actions in W; only that even if it were the case that God were inactive in W, most theists would still claim that He is wholly good in W. I am indebted to Steven J. Jensen for bringing this point to my attention.Google Scholar

¹⁰ It is generally agreed that what it means for God to be necessarily good is that it is impossible (in the broadly logical sense) for God to do evil. Alternatively, one might say that God is necessarily good if and only if God does not do evil in any possible world.Google Scholar

¹¹ I am not implying that it a possible for a being like Clarence to exist; I am merely introducing him here by assumption in order to exploit his unique ontological and moral structure for purposes of comparison with the ontological and moral structure of God. Like any other assumption, when I have finished with him I will discharge him appropriately.Google Scholar

¹² S* is a proper superset of S = df, S is a proper subset of S*. Of course, S is a proper subset of S* just in case S is a subset of S* and S is not identical to S*.Google Scholar

¹³ Of course, the reason for speaking of action-types rather than actions simpliciter (or action-tokens) is that (as a function of Leibniz's Law) for any action A_i (where i ranges over the natural numbers), if G performs A¹ and if G* performs A² (where G and G* are distinct agents), then A¹ is not identical to A². Because A¹ and A² are performed by different agents, A¹ and A² are different action-tokens. But the type of action performed by G and the type of action performed by G* may be such that is identical to For example, consider the action-type: ‘moving the dial 3° counterclockwise’. Suppose G exists in possible world M and G* exists in possible world N. Then, although ‘G's moving the dial 3° counterclockwise’ is not an identical action-token to ‘G*'s moving the dial 3° counterclockwise’, the action-types (i.e. the actions abstracted from the actors) are identical.Google Scholar

¹⁴ S's refraining from performing E in no way implies that S was tempted to performGoogle Scholar

¹⁵ Whatever an ‘action’ is for God, it seems uncontroversial to note that for any one of God's actions A_i, God intends A,. Of course, contrary to the majority of contemporary action theorists (including R. M. Chisholm) this may also be true for us.Google Scholar

¹⁶ Although it seems clear that God (or any other being) may be able to intend that which is necessary in the broadly logical sense (e.g. a necessarily good being can intend to keep all of his promises) it doesn't seem at all clear that what follows from this is that such a being can also intend not to break his promises (or, alternatively, that he could intend to refrain from breaking his promises). What would it even mean to say that such a being intends not to do something he cannot possibly do and that he knows he cannot possibly do? What exactly would he be intending? Could I, for example, intend to be a rock (not merely to have a rock-body, but to actually be a rock) or to actualize any other inconsistent state of affairs? How might I go about intending such things? What could it possibly mean to intend such things?Google Scholar

¹⁷ See, for example, R. M. Chisholm (1979) and A. Donagan (1977) for a spirited discussion on the topic of how to count actions.Google Scholar

¹⁸ The power and versatility of principle (P-3′) lies not only in its relative indifference toward one's manner of individuating actions and action-parts; it can also (as alluded to in note 1) accommodate diverse conceptions of divine goodness. Suppose, for example, that G and G* were both posited to be ‘awholly good’ in some Aristotelian virtue-intensive sense. Given this conception of goodness, it would still be the case that if both G and G* were ‘maximally virtuous’ and (P-3′) (ii) were to hold, then G* would be more benevolent than G.Google Scholar

¹⁹ This paper was presented at the Society of Christian Philosophers Midwest Regional Meeting at the University of Nebraska, Lincoln on 27 October 1990. Travel to this conference was partially funded by Indiana University, South Bend.Google Scholar

²⁰ I would like to thank Robin Collins, Scott A. Davison, John O'Callaghan, and, especially, Thomas P. Flint for valuable comments on earlier drafts of this paper.Google ScholarAdams, R. M. (1972). Must God create the best? Philosophical Review 81, 317–32.CrossRefGoogle ScholarChisholm, R. M. (1979). Person and Object. LaSalle, Ill.: Open Court.Google ScholarDonagan, A. (1977). Chisholm's theory of agency. Journal of Philosophy, 74, 692–703.CrossRefGoogle ScholarGulesarian, T. (1983). God and possible worlds: the modal problem of evil. Nous, (17 05), pp. 221–38.CrossRefGoogle ScholarMorris, T. V. (1987). Anselmian Explorations. Notre Dame: University of Notre Dame Press, pp. 42–69.Google ScholarPike, N. (1969). Omnipotence and God's ability to sin. American Philosophical Quarterly, 6, 208–216.Google ScholarPlantinga, A. (1974). God, Freedom, and Evil. Grand Rapids: Eerdman's Press.Google Scholar